Re: xorg 6.9.0 and DRI
On Wed, Feb 22, 2006 at 01:33:25AM +0100, Michel D�nzer wrote:
>On Tue, 2006-02-21 at 21:34 +0100, Yves-Alexis Perez wrote:
>> On Tue, 2006-02-21 at 14:29 +0100, Michel Dänzer wrote:
>> > On Tue, 2006-02-21 at 08:25 -0500, Brian Victor wrote:
>> > >
>> > > % dpkg -S $(ldd $(which glxinfo)|grep libGL.so|cut -c16-|cut -d' ' -f1)
>> > > xlibmesa-gl: /usr/X11R6/lib/libGL.so.1
>> > Hmm, that looks fine... no idea what's up.
>> For my radeon 9600 I need libgl1-mesa-glx and not xlibmesa-gl
>That's because xlibmesa-dri doesn't have the r300 driver. It doesn't
>explain why xlibmesa-{gl,dri} aren't working for him, but
>libgl1-mesa-{glx,dri} might be worth a try anyway.
Well, I'm not sure how it happened, but it looks like my libGL.so.1
wasn't the right copy. Maybe I did something wacky when I compiled a
prerelease of Xorg, but regardless, I moved that out of the way and DRI
seems to be working now. Thanks for your help!
--
Brian
--
To UNSUBSCRIBE, email to [EMAIL PROTECTED]
with a subject of "unsubscribe". Trouble? Contact [EMAIL PROTECTED]
Re: xorg 6.9.0 and DRI
On Tue, Feb 21, 2006 at 10:42:00AM +0100, Michel D�nzer wrote: >On Mon, 2006-02-20 at 23:52 -0500, Brian Victor wrote: >> % LIBGL_DEBUG=verbose glxinfo >> name of display: :0.0 >> libGL: XF86DRIGetClientDriverName: 4.0.4 r128 (screen 0) >> libGL: OpenDriver: trying /usr/X11R6/lib/modules/dri/r128_dri.so >> libGL error: dlopen /usr/X11R6/lib/modules/dri/r128_dri.so failed >> (/usr/X11R6/lib/modules/dri/r128_dri.so: undefined symbol: >> _glapi_add_dispatch) >What does > >dpkg -S $(ldd $(which glxinfo)|grep libGL.so|cut -c16-|cut -d' ' -f1) > >say? % dpkg -S $(ldd $(which glxinfo)|grep libGL.so|cut -c16-|cut -d' ' -f1) xlibmesa-gl: /usr/X11R6/lib/libGL.so.1 -- Brian -- To UNSUBSCRIBE, email to [EMAIL PROTECTED] with a subject of "unsubscribe". Trouble? Contact [EMAIL PROTECTED]
Re: xorg 6.9.0 and DRI
On Mon, Feb 20, 2006 at 03:36:11PM +0100, Michel D�nzer wrote: >On Sun, 2006-02-19 at 23:46 -0500, Brian Victor wrote: >> When I strace glxinfo, >BTW, you can just set LIBGL_DEBUG=verbose to debug problems like this, >no need to mess with strace. Thanks for the tip! >> it shows that /usr/X11R6/lib/modules/dri/r128_dri.so is sought but not >> found. >You need to install xlibmesa-dri or libgl1-mesa-dri. I installed the former: ii xlibmesa-dri6.9.0.dfsg.1-4 And now I get: % LIBGL_DEBUG=verbose glxinfo name of display: :0.0 libGL: XF86DRIGetClientDriverName: 4.0.4 r128 (screen 0) libGL: OpenDriver: trying /usr/X11R6/lib/modules/dri/r128_dri.so libGL error: dlopen /usr/X11R6/lib/modules/dri/r128_dri.so failed (/usr/X11R6/lib/modules/dri/r128_dri.so: undefined symbol: _glapi_add_dispatch) libGL error: unable to find driver: r128_dri.so display: :0 screen: 0 direct rendering: No -- Brian -- To UNSUBSCRIBE, email to [EMAIL PROTECTED] with a subject of "unsubscribe". Trouble? Contact [EMAIL PROTECTED]
Re: xorg 6.9.0 and DRI
On Thu, Feb 02, 2006 at 06:59:24PM +0100, Michel D�nzer wrote: >On Thu, 2006-02-02 at 18:34 +0100, CK wrote: >> on my tibook IV I kind of lost 3d accelleration since I upgraded xorg >> to 6.9.0 ... >Remove xlibmesa-gl1-dri-trunk (and all the other dri-trunk packages, for >that matter). I have also lost DRI with 6.9. I have a pismo (r128). The log file (http://brianhv.org/temp/Xorg.0.log) shows direct rendering as being enabled, but glxinfo says it is not. I no longer have any dri-trunk packages installed. When I strace glxinfo, it shows that /usr/X11R6/lib/modules/dri/r128_dri.so is sought but not found. This file seems to only exist in xserver-xfree86. Config file is at http://brianhv.org/temp/xorg.conf Any ideas? Thanks! -- Brian -- To UNSUBSCRIBE, email to [EMAIL PROTECTED] with a subject of "unsubscribe". Trouble? Contact [EMAIL PROTECTED]
Pismo External Display
(Paging Michel Daenzer?) I grabbed dri-trunk-unstable in hope that external monitor (unmirrored) support for the r128 in it had found its way in. So far I've only been met with signal 7 crashes. Is support supposed to be in there already? If not, I'm willing to write and test code if someone points me in the right direction. If there is a more appropriate forum for this, please let me know. Config and log: http://brianhv.org/XF86Config-4 (modified from iBook wiki) http://brianhv.org/XFree86.1.log -- Brian
Re: External monitor on iBook rev 2.2
On Sun, Jul 04, 2004 at 08:07:19PM +0200, Cedric Pradalier wrote: >You'll find data file and example on : >http://mockenhaupt.homelinux.org/ibookWiki/Configuration > >I'll try to do a short summary: >With default X server, you can manage this with a two headed >setup, either in xinerama or in two display mode. In both case, with >use fbdev, the behavior of the external output is undefined. So the idea >is to defined two cards, one with fbdev and one without. If you want to >clone your display, you will need to ask xinerama to superpose two >display. In any case, you will loose DRI on both heads. This looks very interesting. Can it, with the RANDR extension, be switched between single-head with DRI and dual-head without at runtime? Does the dual-head work with the Powerbook G3? (I'd go ahead and try it, but at the moment I'm waiting for a replacement hard drive to come in.) -- Brian
Pismo Multihead (again)
Once again I have a use for multihead on my pismo. And once again, X political changes have moved the discussion venues. Should I be using [EMAIL PROTECTED] to ask about things like this now? At any rate, what's going on with multihead on the pismo? Last I heard (quite a while ago) someone on Xpert had very limited success that wasn't widely reproduced. Has anything else emerged since then? Is there any code I can write once I free up some time in a couple of weeks? Should I just be asking all this on fd.o? Thanks, -- Brian
Pismo question (was Re: powerbook 15" 1Ghz question ...)
On Sat, Apr 19, 2003 at 01:31:30AM +0200, Michel D?nzer wrote: >On Sam, 2003-04-19 at 01:53, Thomas Bodenmann wrote: >> Michel D?nzer <[EMAIL PROTECTED]> writes: >> > It did last I tried, that was with the M7, but I don't expect any change >> > with the M9 there. >> Mirroring or Xinerama? >Anything. It's time for my semi-annual question. ;) What's the progress on the Pismo multihead code? Last time I asked you directed me to the xpert list, but I can't seem to find that list anymore. At that point, the code was still in a "may set your display on fire" state. Thanks! -- Brian
Re: tar ate my symlinks
On Mon, Mar 17, 2003 at 09:43:09AM -0800, Vineet Kumar wrote: >* Brian Victor <[EMAIL PROTECTED]> [20030317 07:52 PST]: >> I backed up my debian installation with the following: >> >> tar --preserve -cv / | ssh 192.168.2.10 'cat > linuxbackup.tar.bz2' > >That .bz2 extension on that file is misleading. The usual extension for >uncompressed tar archives (which is what this line creates) is .tar . Oops, s/cat/bzip2/, as you figured out. >> Both commands were run as root. However, I now have a *lot* of broken >> symlinks, which show up like this: >> --1 root root0 Mar 17 13:19 xdvi.bin >> --1 root root0 Mar 17 13:19 xmlproc_parse >> --1 root root0 Mar 17 13:19 xmlproc_val >> --1 root root0 Mar 17 13:19 yacc >Very strange. What type of filesystem are you running on? Any peculiar >mount options? /dev/hda4 on / type ext3 (rw,errors=remount-ro) >Can you tell if the tarball actually has the symlinks in it but the >restore messed up, or if the files are stored incorrectly and restored >correctly? I don't know; is there any way to tell that in a compressed (or even a decompressed) tarball? >Then, in order to preserve them, you'll probably need another debian >installation to figure out what they're supposed to be. Another thing >you could try is to reinstall all of your packages. I would guess that >this would take care of most of them. Yes, but it's difficult when some fundamental programs like awk, csh, and yacc are broken. It's hard to tell what needs to be reinstalled first. Also, as far as I can tell, I have to uninstall before installing again, which can cause dependency messes. I do have another debian box in a similar state, so the manual method may indeed be the best. Thanks! -- Brian
Re: Getting my Alt key back
On Fri, Dec 13, 2002 at 07:35:44PM +0100, Michel D?nzer wrote: >On Fre, 2002-12-13 at 02:50, Brian Victor wrote: >> I recently switched from WindowMaker to evilwm, and in the process, lost >> the swapmeta functionality. >I think an X upgrade did that, haven't investigated further. OK, but if I switch back to WindowMaker with the same version of X (4.1.0-16), the swapmeta works again. That leads me to believe it's not isolated to an X version issue. -- Brian
Getting my Alt key back
I recently switched from WindowMaker to evilwm, and in the process, lost
the swapmeta functionality. Is there anything I can do (including
hacking evilwm, if necessary) to make the command key my alt key again?
Portion of XF86Config-4:
Section "InputDevice"
Identifier "Generic Keyboard"
Driver "keyboard"
Option "CoreKeyboard"
Option "XkbRules" "xfree86"
Option "XkbModel" "macintosh"
Option "XkbLayout" "us"
Option "XkbOptions""alt:swapmeta"
EndSection
Portion of /etc/X11/xkb/symbols/alt:
// swap the alt and "windows" keys
partial modifier_keys
xkb_symbols "swapmeta" {
key { symbols[Group1] = [ Meta_L ] };
key { symbols[Group1] = [ Alt_L ] };
key { symbols[Group1] = [ Meta_R ] };
key { symbols[Group1] = [ Alt_R ] };
};
--
Brian
Revisiting Pismo Multihead
Given the recent update to the ATI Linux drivers, I was wondering if any more information was available to allow multihead operation on the Pismo. My guess is no, since the new drivers seem to be x86 only and for newer cards, but I thought something else might have happened since the summer when I last asked. Would development effort for this be best supported through the DRI project, XF86, or elsewhere? Or am I even on the wrong list to be asking this question? If anyone can give me new information on this, I'd appreciate it. I'd love to be rid of the last reason I have for booting OSX. -- Brian
Re: mozilla problems
On Fri, Sep 20, 2002 at 10:14:07PM -0400, Chris Hoover wrote: >I'm having problems with mozilla. It now longer will fully launch. >When I run it from a xterm or from a menu, it starts up but I never get >the splash screen. The program never seems to die, and it does not give >me any errors. killall esd Newer versions of mozilla want to open the sound device for some reason, and they'll wait forever until they get it. Killing esd or whatever is using the sound device will free it up so that it opens. Once up, you can restart esd and mozilla will keep working. esddsp mozilla may also work; I haven't tried it, and I'm not in linux at the moment. -- Brian "Our greatest glory consists Email: [EMAIL PROTECTED] not in never falling, but in Web: http://www.brianhv.org/ rising every time we fall." -- Oliver Goldsmith
usb-ohci hard lock
I have a reproduceable condition which is producing a hard lockup (no keyboard input even at console, no network activity) consistently. I installed a Belkin BusPort USB card (listed as supported at penguinppc.org) and installed the usb-ohci driver as a module. I attached a Samsung ML-1210 printer to it, with CUPS 1.1.14-3 and the Samsung PPC Linux drivers installed. Using the samsung-config program, I try to print a test page. At this point, the job is queued and the job is at least partially sent to the printer, as indicated by the blinking "data" light. However, before the job completes, the system locks up. This happened with 2.4.17, and still happens with an rsync'ed 2.4.19-rc1-ben0 from this morning, with usb-ohci and printer both running as modules. The crash occurs on my beige G3/233 desktop. The printer works successfully with a similar configuration (except with lpr-ppd instead of CUPS) on my pismo. Please let me know if I can provide additional information that can help in debugging this problem. -- Brian "I am an optimist. It does Email: [EMAIL PROTECTED] not seem too much use being Web: http://www.personal.psu.edu/bhv1/ anything else." -- Winston Churchill -- To UNSUBSCRIBE, email to [EMAIL PROTECTED] with a subject of "unsubscribe". Trouble? Contact [EMAIL PROTECTED]
gsfonts-other incomplete install
In attempting to install gsfonts-other from woody, apt reports the following: %apt-get install Reading Package Lists... Done Building Dependency Tree... Done 0 packages upgraded, 0 newly installed, 0 to remove and 409 not upgraded. 1 packages not fully installed or removed. Need to get 0B of archives. After unpacking 0B will be used. Setting up gsfonts-other (6.0-3) ... (Re-)registering Postscript fonts.../usr/bin/defoma-font: /etc/defoma/hints/gsfonts-other.hints: syntax error in line 61. 'category' after 'begin'. dpkg: error processing gsfonts-other (--configure): subprocess post-installation script returned error exit status 1 Errors were encountered while processing: gsfonts-other E: Sub-process /usr/bin/dpkg returned an error code (1) The file in question has a "category" line in line 61, though it is not within a begin/end block. It is, however, the only file within that directory with more than one category line. I'm not familiar with the use or syntax of those files to know how to fix it. Any ideas? -- Brian "Whoever has a heart full of Email: [EMAIL PROTECTED] love always has something Web: http://www.personal.psu.edu/bhv1/ to give." -- Pope John XXIII -- To UNSUBSCRIBE, email to [EMAIL PROTECTED] with a subject of "unsubscribe". Trouble? Contact [EMAIL PROTECTED]
Re: Sound on a PowerBook G3 Firewire
On Tue, Jun 25, 2002 at 09:50:12AM +0900, [EMAIL PROTECTED] wrote: >>>>>> "Brian" == Brian Victor <[EMAIL PROTECTED]> writes: >Brian> I have never gotten the built-in mic or the built-in line >Brian> input to be useful under linux on my Powerbook G3. I did >Brian> pick up a Griffin iMic (USB mic input), and have found some >Brian> success with it. It's been a couple of months since I used >Brian> it, but I recall needing to explicitly specify a recording >Brian> speed; otherwise, I got clicks in the audio. >Do you have the details on how you managed to get it working? I have >a Logitech (Phillips) webcam with the built in mic, and would like to >get that working, if possible. Keep in mind that it's been a little while since I did it, but here's what I remember. Step 1: make sure that USB audio is available to the kernel, probably as a module. (I'm assuming your webcam is USB. I'm also assuming it registers itself as an audio device, which may be completely wrong.) Step 2: check your audio mixer of choice (I use gmix) to see if your audio device is being recognized. gmix displayed a new tab for me, something like "USB Audio Class Device". Step 3: Use the following command: rec -d /dev/dsp1 -r 44100 myrecording.wav /dev/dsp1 is the second audio device (after builtin, /dev/dsp). 44100 is, IIRC, CD-quality sound, and also (I think) the max supported by the iMic. I would presume that a more comprehensive audio recording program would also find success with those bitrate and device settings. You may have to play around with the bitrate to get something acceptable. Also, note that I was using OSS. I haven't managed to get ALSA to do much for me. -- Brian "We are each of us angels with Email: [EMAIL PROTECTED] only one wing, and we can fly Web: http://www.personal.psu.edu/bhv1/ only by embracing each other." -- Lucian de Crescenzo -- To UNSUBSCRIBE, email to [EMAIL PROTECTED] with a subject of "unsubscribe". Trouble? Contact [EMAIL PROTECTED]
Re: Sound on a PowerBook G3 Firewire
On Tue, Jun 25, 2002 at 01:22:33AM +0900, [EMAIL PROTECTED] wrote: >Has anyone been able to get sound recording working on the PowerBook >G3 Firewire? Using Gnomemixer, I can turn up the line in for the >build in mic, and hear the input from it, but when I try to use some >program like audicity, I can't seem to point it at anything that will >get me some useful sound. Not ever a screech. I have never gotten the built-in mic or the built-in line input to be useful under linux on my Powerbook G3. I did pick up a Griffin iMic (USB mic input), and have found some success with it. It's been a couple of months since I used it, but I recall needing to explicitly specify a recording speed; otherwise, I got clicks in the audio. I'm sure this isn't quite what you wanted to hear, but I hope it's at least a little helpful. -- Brian "One's friends are that part of Email: [EMAIL PROTECTED] the human race with which one Web: http://www.personal.psu.edu/bhv1/ can be human." -- George Santayana -- To UNSUBSCRIBE, email to [EMAIL PROTECTED] with a subject of "unsubscribe". Trouble? Contact [EMAIL PROTECTED]
Fixing Broken System
Due to an intermittent hard drive error which has since been fixed, my root partition has taken significant but not sweeping damage. Most binaries still work, but a few don't, such as df, cat, and mount. I don't know precisely how much was damaged, nor do I know an easy method of determining this. That said, is there any way of forcing apt to reinstall all installed packages or something? From what I can tell, upgrade and dist-upgrade only reinstall changed packages. Or perhaps I can install the potato base and use my existing apt database to tell it what packages to install? What is the recommended way of recovering such a broken system? Thanks, -- Brian "The only time you should look Email: [EMAIL PROTECTED] down on a man is when you're Web: http://www.personal.psu.edu/bhv1/ helping him get up." -- Unknown -- To UNSUBSCRIBE, email to [EMAIL PROTECTED] with a subject of "unsubscribe". Trouble? Contact [EMAIL PROTECTED]
Re: Slow Man
On Sun, May 19, 2002 at 09:35:05PM -0500, Colin Watson wrote: >On Mon, Apr 01, 2002 at 10:40:06PM -0500, Brian Victor wrote: >> My "man" command takes an inordinately long time to run: [snip] >If /var/cache/man/oldlocal/index.bt is still zero-length, then you can >work around this for now by just removing it. It and several other index.bt files were indeed zero-length. Removing them gives me a nice fast man command again. Thanks! -- Brian "The only power which can resist Email: [EMAIL PROTECTED] the power of fear is the power Web: http://www.personal.psu.edu/bhv1/ of love." -- Alan Paton -- To UNSUBSCRIBE, email to [EMAIL PROTECTED] with a subject of "unsubscribe". Trouble? Contact [EMAIL PROTECTED]
Re: airport eth1, onboard eth0 & eth1 Tx error, status 1
On Fri, May 17, 2002 at 02:40:10PM -0400, K Clark wrote: >onboard eth0 is working flawlessly. have used dhclient-2.2.x and pump, both >without problems. > >airport worked ONCE last evening. so here's the dilemma, and how i got there. Someone more informed may be able to give you a better answer, but I've only been able to get my airport to work when my onboard ethernet was down. Whenever I want to connect via airport, I do ifdown eth0; ifup eth1, and vice-versa. The reason for this seems to be that Linux can't figure out which device to route packets to when both interfaces are up. If you ifup both interfaces and look at the route command, you can see similar lines for both interfaces. I presume this is the cause of the problems. There's probably a way around it with some combination of the route and ifconfig commands, but it hasn't been worth it for me to figure it out. -- Brian "We are each of us angels with Email: [EMAIL PROTECTED] only one wing, and we can fly Web: http://www.personal.psu.edu/bhv1/ only by embracing each other." -- Lucian de Crescenzo -- To UNSUBSCRIBE, email to [EMAIL PROTECTED] with a subject of "unsubscribe". Trouble? Contact [EMAIL PROTECTED]
Re: windowmaker key combo. to switch workspaces, et al
On Fri, May 17, 2002 at 06:53:41AM -0400, K Clark wrote: >i haven't figured out the key combo to switch workspaces, or jump out of X to >another tty. Someone else answered the TTY question, so I'll cover windowmaker. The key combinations are defined in the WPrefs app, which I believe has an icon on the dock by default. If you run WPrefs and go to the icon with the keyboard on it, you can view and define key combinations. My setup uses alt-1 through alt-5 to switch workspaces, as well as ctl-arrow keys. -- Brian "One's friends are that part of Email: [EMAIL PROTECTED] the human race with which one Web: http://www.personal.psu.edu/bhv1/ can be human." -- George Santayana -- To UNSUBSCRIBE, email to [EMAIL PROTECTED] with a subject of "unsubscribe". Trouble? Contact [EMAIL PROTECTED]
Re: XFree86 and xinerama on ppc
On Thu, May 16, 2002 at 09:28:41AM +0200, Michel D?nzer wrote: >On Tue, 2002-05-14 at 23:01, Brian Victor wrote: >> Do the changes need to be made to the kernel module, the XFree module, >> or both? >Depends what kernel module you mean, and what you want to use it for. Basically I wanted to know if the kernel needed to initialize the CRT before X could use it. From looking at the X sources, though, that seems to be unneccessary. >For X, check out the IsSecondary flag in >programs/Xserver/hw/xfree86/drivers/ati/radeon*.[ch] (4.2 or current >CVS). That helps somewhat, in that I see basically what's going on once the variables get set up. But getting the structures defined and initialized seems like it'll be the biggest challenge. There are two giant ATI patches in CVS dealing with multi-head on the Radeon. The first was a 1716 line patch that enabled support, and it was followed three weeks ago with a 2232 line clean-up patch. It looks like a whole lot of both of those patches needs to be ported over to r128 to get this to go. And thats just in the driver files alone. IOW, this isn't looking like a one-weekend project. Unless I'm missing some shortcut. I probably won't be able to make a serious run at this until July at the very earliest, since I have a thesis and other such things demanding my time until then. So if anyone else wants to start hacking away at it, be my guest. -- Brian "We are each of us angels with Email: [EMAIL PROTECTED] only one wing, and we can fly Web: http://www.personal.psu.edu/bhv1/ only by embracing each other." -- Lucian de Crescenzo -- To UNSUBSCRIBE, email to [EMAIL PROTECTED] with a subject of "unsubscribe". Trouble? Contact [EMAIL PROTECTED]
Re: XFree86 and xinerama on ppc
On Tue, May 14, 2002 at 09:26:30PM +0200, Michel D?nzer wrote: >On Tue, 2002-05-14 at 18:37, Brian D. Hicks wrote: >> Speaking of powerbook dualhead, I don't suppose any progress has been >> made on xinerama support for the Rage 128 mobility in the Pismo? >I'm not aware of any changes there. Porting that support from the radeon >driver would be a relatively easy project for someone who'd like to try >some X hacking. :) Can you tell me what I'm looking for or point me to a doc or mailing list? I'm not making any promises, but I'd really like to see this working, and a little X hacking never hurt anyone. (Right? ;) Do the changes need to be made to the kernel module, the XFree module, or both? Thanks, -- Brian "You make a living out of what Email: [EMAIL PROTECTED] you get. You make a life out Web: http://www.personal.psu.edu/bhv1/ of what you give." -- Unknown -- To UNSUBSCRIBE, email to [EMAIL PROTECTED] with a subject of "unsubscribe". Trouble? Contact [EMAIL PROTECTED]
Re: Locale Settings
On Mon, May 06, 2002 at 02:31:01AM +0100, Bastien Nocera wrote: >Run "dpkg-reconfigure locales" and make sure that fr_FR is one of the >locales you enabled and generated, otherwise it will silently fallback >to the C locale. That fixed it. Thanks! -- Brian "Our greatest glory consists Email: [EMAIL PROTECTED] not in never falling, but in Web: http://www.personal.psu.edu/bhv1/ rising every time we fall." -- Oliver Goldsmith -- To UNSUBSCRIBE, email to [EMAIL PROTECTED] with a subject of "unsubscribe". Trouble? Contact [EMAIL PROTECTED]
Locale Settings
I'm attempting to use translations for some programs, but have been unsuccessful. I've tried various settings for LC_ALL and LANG, including fr, fr_FR, and french, but gedit doesn't use the translated messages. /usr/share/locale/fr/LC_MESSAGES/gedit.mo exists. I have the gettext and locales packages installed. What am I missing? Thanks, -- Brian "I am an optimist. It does Email: [EMAIL PROTECTED] not seem too much use being Web: http://www.personal.psu.edu/bhv1/ anything else." -- Winston Churchill -- To UNSUBSCRIBE, email to [EMAIL PROTECTED] with a subject of "unsubscribe". Trouble? Contact [EMAIL PROTECTED]
Re: Slow Man
On Wed, Apr 03, 2002 at 03:08:59PM -0600, Colin Watson wrote: >Hm. Adam suggested that it could be a kernel 2.2 bug, if you're running >that, pointing to the various hits you get for '_newselect powerpc' on >google. I'm already way out of my league on powerpc-specific problems, >though. I'm running kernel 2.4.17-ben0. However, it looks like all the discussions on google were about oldworld machines, and I'm no different; the problem is occurring on my oldworld beige G3, but not my newworld pismo. I did not have this problem with my old LinuxPPC distribution, for whatever that's worth. Should I be reporting this information to a bug tracking system? -- Brian "Nothing in this life is to be Email: [EMAIL PROTECTED] feared. It is only to be Web: http://www.personal.psu.edu/bhv1/ understood." -- Marie Curie -- To UNSUBSCRIBE, email to [EMAIL PROTECTED] with a subject of "unsubscribe". Trouble? Contact [EMAIL PROTECTED]
Re: Slow Man
On Wed, Apr 03, 2002 at 05:34:27AM -0600, Colin Watson wrote: >On Wed, Apr 03, 2002 at 05:16:40AM -0600, Colin Watson wrote: >> This looks like it might be #137908. I'd appreciate help on this - it >> seems to be powerpc-specific. >I've tried it on a powerpc now (thanks, Adam Conrad) and couldn't >reproduce the problem there either. In fact, on that machine I don't see >any _newselect() calls at all. > >What does 'ls -l /usr/lib/man-db/man' say, and what version of libdb2 do >you have installed? %ls -l /usr/lib/man-db/man -rwsr-xr-x1 man root 103844 Mar 15 17:03 /usr/lib/man-db/man libdb2 = 2.7.7.0-2. My pismo, whose man pages work fine, have man installed as the root user, but not setuid. It's using libdb2 2.7.7.0-1. Changing the permissions to those on the pismo have no change to the _newselect problem. Updating to libdb2 2.7.7.0-5 does not seem to change anything, either. -- Brian "One's friends are that part of Email: [EMAIL PROTECTED] the human race with which one Web: http://www.personal.psu.edu/bhv1/ can be human." -- George Santayana -- To UNSUBSCRIBE, email to [EMAIL PROTECTED] with a subject of "unsubscribe". Trouble? Contact [EMAIL PROTECTED]
Slow Man
My "man" command takes an inordinately long time to run:
%time man foo
No manual entry for foo
man foo 0.01s user 0.02s system 0% cpu 21.026 total
strace shows the process pausing for almost a full second at each
_newselect() call (the below occurs 21 times for a nonexistent man page):
open("/var/cache/man/oldlocal/index.bt", O_RDONLY) = 3
fcntl64(3, F_SETFD, FD_CLOEXEC) = 0
fstat(3, {st_mode=S_IFREG|0644, st_size=0, ...}) = 0
read(3, "", 512)= 0
close(3)= 0
_newselect(0, 0, 0, 0, 0x7488) = 0
Any ideas about what might be happening? I have the following installed:
||/ Name VersionDescription
+++-==-==-
ii man-db 2.3.20-18 The on-line manual pager
ii manpages 1.39-1.1 Man pages about using a Linux system.
ii manpages-dev 1.29-2 Linux-development man pages.
--
Brian
"I am an optimist. It does
Email: [EMAIL PROTECTED] not seem too much use being
Web: http://www.personal.psu.edu/bhv1/ anything else."
-- Winston Churchill
--
To UNSUBSCRIBE, email to [EMAIL PROTECTED]
with a subject of "unsubscribe". Trouble? Contact [EMAIL PROTECTED]
G4 Upgrade
I'm looking at finally taking the plunge and doubling the speed of my beige G3 by popping a G4 ZIF in. OWC says that the ZIF is linux-compatible. However, I have two questions: 1) Does the G4 have any advantage in linux over a G3? Last I knew, gcc didn't compile for Altivec, though I thought I recalled hearing that Apple was merging their Altivec compiling code. (OT - can anyone tell me how I'd be able to code to take advantage of Altivec if gcc *did* compile for it?) 2) The ZIF uses cache-enabling software under MacOS. Is this something that can be run once under MacOS then work under linux? If not, is it something that can run before BootX and then work under linux? Thanks, -- Brian "Nothing in this life is to be Email: [EMAIL PROTECTED] feared. It is only to be Web: http://www.personal.psu.edu/bhv1/ understood." -- Marie Curie -- To UNSUBSCRIBE, email to [EMAIL PROTECTED] with a subject of "unsubscribe". Trouble? Contact [EMAIL PROTECTED]
Re: How to switch between virtual terminals on PowerBook?
On Wed, Mar 20, 2002 at 07:46:47PM +0100, Vincent Lefevre wrote: >On Wed, Mar 20, 2002 at 18:31:22 +, Clive Menzies wrote: >> (held together) + works on my 8100/80. >Here it does nothing. :( It seems that my key isn't taken into >account, whatever I do. I'm kind of surprised no one has mentioned this before, but have you tried the aforementioned keystrokes with [left] option instead of the apple key? For my pismo, the apple key seems worthless, and I have to use the option key for meta. On my pismo, opt-fn-F* switches terminals from console mode, as does opt-left/right. From X, opt-ctl-fn-F* does the trick. -- Brian "The only time you should look Email: [EMAIL PROTECTED] down on a man is when you're Web: http://www.personal.psu.edu/bhv1/ helping him get up." -- Unknown
Re: searching Java & flash for mozilla
On Fri, Jan 25, 2002 at 05:02:59PM -0800, Wilhelm *Rafial* Fitzpatrick wrote: >At 4:45 PM -0500 1/25/02, Adam Goode wrote: >>Java can be found (in various forms) as part of Debian, and as part of >>the Blackdown project (http://www.blackdown.org/java-linux.html). The >>Blackdown version contains Sun's Java plugin which should work on >>Mozilla. >Sadly, the PPC java plugin (which can be found at >http://penguinppc.org/projects/java) broke a couple Mozilla revs >(around 9.5 or so) back, and no one to my knowledge has managed to >release an updated version. > >If I am incorrect about this, please let me know, I sorely miss Java >when surfing with Mozilla. Actually, blackdown released 1.3.1-02a-FCS three days ago. I installed it as soon as I found out and java has been working about as well as I can expect in Mozilla 0.9.7. (i.e., most applets work pretty well, a few like to munch processor time eternally.) See the address Adam gave for details. -- Brian "The only power which can resist Email: [EMAIL PROTECTED] the power of fear is the power Web: http://www.personal.psu.edu/bhv1/ of love." -- Alan Paton
Re: deja vu: problems with latex and accents
On Sat, Jan 19, 2002 at 06:42:21PM +, Rory Campbell-Lange wrote:
>There are two accents in the word deja, é and an à which I can see in
>downloaded text (in an xterm) but cannot find the key combination for.
>I'm using woody on a Lombard Powerbook.
>
>How do I get latex to 'render' these in the dvi output? At the moment
>they aren't rendered at all. Déjà vu becomes Dj vu!
LaTeX uses escapes to render such characters. Use:
d\'{e}j\`{a}
in your .tex source file.
--
Brian
"The only power which can resist
Email: [EMAIL PROTECTED] the power of fear is the
power
Web: http://www.personal.psu.edu/bhv1/ of love."
-- Alan Paton
Re: Ogg Vorbis Hissing
On Mon, Jan 07, 2002 at 11:23:25AM -0200, Rog?rio Brito wrote: >On Jan 06 2002, Brian Victor wrote: >> Ogg Vorbis files played through xmms on my beige G3 result in static. [snip] >The maintainer of the package has told me that it will be fixed in the >next version of xmms. Indeed. I downloaded 1.2.6-pre2 and installed from source, and my ogg files play fine now. Thanks for pointing that out. -- Brian "Nothing in this life is to be Email: [EMAIL PROTECTED] feared. It is only to be Web: http://www.personal.psu.edu/bhv1/ understood." -- Marie Curie
Ogg Vorbis Hissing
Ogg Vorbis files played through xmms on my beige G3 result in static. A nearly identical setup on my pismo works fine. ogg123 works fine on both machines. This is with xmms 1.2.5. Both machines are running ben0 kernels; the desktop has 2.4.17, while the pismo has 2.4.15. Is there a known fix for this? Thanks, -- Brian "The only time you should look Email: [EMAIL PROTECTED] down on a man is when you're Web: http://www.personal.psu.edu/bhv1/ helping him get up." -- Unknown
Re: 3d support on pismo
On Sun, Nov 25, 2001 at 12:48:56AM +0100, Michel Dänzer wrote: >Read the documentation on http://dri.sf.net, in particular the DRI >Beginner's Guide and the DRI User Guide. I'm not the original poster, but the question and response were timely. Unfortunately, I'm still hitting some problems. In 15 and 24 bits on my pismo, DRI doesn't get enabled. In 16 bits, DRI works fabulously, but the colormap is wrong; things are very green. I'm running kernel 2.4.15-pre5. DRI is compiled in, and I have the rage 128 module built. My XF86Config-4 Device section looks like this: Section "Device" Identifier "Generic Video Card" Driver "ati" Option "UseFBDev" BusID "PCI:0:16:0" EndSection I also tried "r128" for the driver, and I've confirmed that the BusID is correct. Any ideas as to how to get normal colors in 16 bits? Thanks, -- Brian "We are each of us angels with Email: [EMAIL PROTECTED] only one wing, and we can fly Web: http://www.personal.psu.edu/bhv1/ only by embracing each other." -- Lucian de Crescenzo
Partition Limit?
I'm trying to triple boot (Mac 9/X, Debian) on my new 30 gig drive. OS X put a bunch of small partitions on the drive, bumping my linux partitions to numbers 14, 15, and 16. Before OS X did so, there were 4 more gigs of free space after the linux partitions. mac-fdisk no longer shows this space. Is this because of a 16-partition limit? Or did OS X hose the partition map somehow? Does anyone know of a workaround? Thanks, -- Brian "We are each of us angels with Email: [EMAIL PROTECTED] only one wing, and we can fly Web: http://www.personal.psu.edu/bhv1/ only by embracing each other." -- Lucian de Crescenzo
Netgear Card in Installer
I'm attempting to migrate from LinuxPPC. I would like to install from network since I have a very local mirror. However, my beige G3's onboard ethernet is having compatibility issues with my campus network, so I need to use my Netgear FA311. Unfortunately, the National Semiconductor drivers aren't in the install kernel/ramdisk. I have natsemi.o from my 2.2.19 LinuxPPC modules directory, but trying to insmod it is met with half a dozen unresolved symbol errors. I was able to succesfully insmod pci-scan.o, the only module on which natsemi.o seems to depend on my old system. What else do I need to do to use the FA311 in the installer? Thanks, -- Brian "A man who wants to lead the Email: [EMAIL PROTECTED] orchestra must turn his back Web: http://www.personal.psu.edu/bhv1/ on the crowd." -- Max Lucado
